ConceptSearch http://www.conceptsearching.com/web/ is a commercial search engine and classification tool. Maybe similar to TemaTres, it doesn't use machine-learning but extracts "concepts" out of your documents that can be mapped to vocabulary terms. The vocabulary is then exposed to the end-user as search results facet. It's all driven by MS SQL Server and exposed as web services. We've used it here to map medical school lectures to the licensing exam outlines and have experimented a little with autoclassifying the same lecture content by MeSH. Jason Jason Stirnaman Biomedical Librarian, Digital Projects A.R. Dykes Library, University of Kansas Medical Center [log in to unmask] 913-588-7319 >>> On 11/28/2011 at 12:00 AM, in message <[log in to unmask]>, Peter Neish <[log in to unmask]> wrote: Hi there, Just wondering if anyone has any recommendations for systems that will do automatic content classification through machine learning? We want to classify newspaper articles using terms from our existing thesaurus and have a fairly big set of articles already tagged that could be used as a training set.. Services like OpenCalais don't really fit our need because we want to use our own thesaurus. Happy to look at both open source and commercial software. Thanks, Peter -- Peter Neish Systems Officer Victorian Parliamentary Library Ph: 03 9651 8638 [log in to unmask] ///////************************************************************/////////////// Parliament of Victoria . Important Disclaimer Notice: The information contained in this email including any attachments, may be confidential and/or privileged. If you are not the intended recipient, please notify the sender and delete it from your system. Any unauthorised disclosure, copying or dissemination of all or part of this email, including any attachments, is not permitted. This email, including any attachments, should be dealt with in accordance with copyright and privacy legislation. Except where otherwise stated, views expressed are those of the individual sender.